Interventions which reduce the need for people to attend clinics are proving highly successful in retaining people in care and supporting adherence to HIV medication in southern Africa, the 21st International AIDS Conference heard last week.
Measures to reduce the burden of people seeking health care are also critical to improving the capacity of health systems to manage growing numbers of patients, numerous presenters at the conference confirmed. The new wave of interventions – described as 'differentiated care' in guidelines – are intended to reduce clinic visits, waiting times and monitoring requirements.
The benefits for patients include less time spent waiting in clinics and travelling to clinics, fewer out-of-pocket travel expenses and less time off work due to clinic attendance, and more support in the community for adherence to medication. The benefits for health services come in the form of increased capacity to deal with growing patient numbers, more time to concentrate on patients with complex needs, and better retention of patients in care due to the use of community health workers and other community-level mechanisms for supporting treatment.
In new guidance issued ahead of the conference, the World Heath Organisation (WHO) urged national treatment programmes to begin thinking in terms of delivering treatment to four different groups of patients, and tailoring services for each group accordingly.
The four groups of patients are: people presenting when well: new patients who will need adherence and retention support as they start treatment, and monitoring during the early months of treatment; people with advanced disease: new patients who present with symptomatic HIV disease or CD4 counts below 200, or who develop TB, who will need fast-track clinical care and more intensive follow-up; stable patients – people on treatment for at least one year with undetectable viral load, not pregnant or breastfeeding. Unstable patients – people on treatment with detectable viral load, who need adherence support, possible second or third-line switches, monitoring for HIV drug resistance; and people can be expected to transition from one group to another – in the majority of cases from the 'Presenting when well' to the 'Stable' category.
"A one size fits all approach to care is no longer suitable," said Gottfried Hirnschall, director of the WHO HIV/AIDS programme, introducing the guidance. A new, differentiated approach to care is needed, said Anna Grimsrud of the International AIDS Society. "We’re not getting the retention in care that we need, so something is wrong for patients. We need to treat all, so we will need to treat more people, and in order to reach the 90-90-90 targets, we need to speed up," she told a pre-conference satellite meeting on differentiated care.
Differentiated care involves not only the shifting of tasks to new cadres in the health system, such as community health workers, but the assumption of responsibility for managing elements of their own care by largely self-organised patient groups. These mechanisms include the distribution of medication by patient groups, which may require changes in rules in many countries.
"One of the biggest barriers to differentiated care is regulatory – rules that say this person cannot do this," said Carlos del Rio of Emory University.
The conference heard findings on a number of models for differentiated care including six-monthly appointments, adherence clubs and community antiretroviral therapy (ART) refill groups.
But Dr Eric Goemaere of Médecins sans Frontières (MSF) warned that community services like adherence clubs are an extra cost. "Clinics still need to run," he said, pointing out adherence clubs need to be understood as a mechanism to expand the volume of patients treated, not as a cost-saving mechanism.
A review of a switch to six-monthly appointments for clinically stable patients in Malawi found that the switch reduced attrition from HIV care and saved 30,000 clinician consultations in one district in 2014 alone. MSF switched from monthly or three-monthly appointments to six-monthly appointments for clinically stable patients in its treatment programme in Chiradzulu district as far back as 2008. Patients obtained drug refills from the pharmacy every three months. The programme provides care for around 35,000 patients, 95% of whom are now on ART.
The analysis looked at outcomes in 24,802 patients in treatment since 2008 who were eligible for six-monthly appointments, of which 18% did not take up the option of less frequent clinic visits. Those who did not enrol were significantly more likely to die or become lost to follow-up (adjusted odds ratio 3.09, 95% CI 2.47-3.87) possibly an indication that they were considered unsuitable for switching to less frequent clinical contact, despite being clinically stable. Overall, only 3% of those who were later enrolled to six-monthly appointments were lost to care, compared to 35% of those never enrolled.
As for appointments saved, the analysis showed that a switch to six-monthly appointments only began to have a substantial impact on the total number of clinician appointments in 2014, the first year in which clinician visits declined substantially. Presenting the results, Alison Wringe of the London School of Hygiene & Tropical Medicine said that rollout of six-monthly appointments had been relatively slow, but enrolment was expected to speed up with the introduction of routine viral load monitoring.
The SEARCH study of community-based testing, expedited linkage to care and treatment for all found that its model of streamlined clinic care resulted in substantially shorter patient visits – on average, around one hour less, and that both reduced waiting time and reduced clinical consultation time explained this difference.
"Nobody likes to wait, and in these communities, patients may wait up to four to five hours to see a clinician for five minutes," said Starley B Shade of University of San Francisco California, a member of the SEARCH study team. Anticipation of such long waiting times may deter patients from attending the clinic, especially if it involves a loss of income.
The SEARCH study used a streamlined care system in its intervention communities in Uganda and Kenya in which nurses carried out triage of patients on arrival at the clinic, directing patients through blood draws, clinician appointments and pharmacy refill visits. The study included a time-and-motion element in which patients were given a form to carry through their clinic visit on which the start and finish times of each encounter with a service provider were recorded. Researchers compared waiting times and entire clinic visit times for intervention clinics (353 patients) and government clinics (745 patients) providing standard of care services to the control arm in the study. They found a mean visit length of 1.08 hours for those with CD4 counts above 500 and 1.13 hours for those with CD4 counts below 500 at intervention clinics, compared to a mean of 2.35 hours at government clinics, of which over two hours was spent waiting, with the longest waiting times for a clinical service and for a pharmacy refill. A quarter of patients at government clinics spent more than 3.5 hours at the clinic.
The streamlined care model freed clinical officers to see patients in need of clinical attention, and also reduced the total number of patient visits each day due to better planning of clinic appointments.
Swaziland has extremely high HIV prevalence (31%) but many people who need ART are still untreated. Expanding the capacity of the health system to provide antiretroviral treatment in this largely rural country will require a shift towards community-based health care. MSF implemented a pilot programme in Swaziland to evaluate the success of moving clinically stable patients on ART to community care models in 2015 and 2016. (Lukhele)
Patients in different types of facilities were given the option of moving to various types of community care, all of which offered a reduced amount of clinic attendance:
Facility-based adherence clubs (three-monthly) where around 30 patients attend to pick up pre-packaged medication. Blood draws for viral load and clinical symptom reporting formed part of each session.
Community ART groups of around six patients, self-formed by patients, at rural facilities, in which members take turns to collect medication and attend the clinic; group sessions the following day involve medication pick-up, pill counts, adherence support and checking weight. Outreach service for very remote areas, where pre-packaged medication is delivered monthly.
By the end of the second quarter of 2016, 727 patients had been enrolled in community ART programmes, 40% in nine adherence clubs, 46% in 60 community ART groups and 14% in three outreach communities.
Patients in each group had median CD4 cell counts above 500 and had been on ART for more than five years. Visit attendance for each mode was very high: 96% for adherence clubs and community outreach and 100% for community ART groups. Forty-one patients returned to mainstream clinic care; patients were significantly more likely to be retained in adherence clubs than in community ART groups or in outreach care.
The main reasons for returning to mainstream care were a lack of eligibility for community care in the first place, viral load test results that required clinical attention, and communication issues within the community ART group.
The MSF evaluation concluded that community management of ART is feasible and that community health workers and lay people have important roles to play in establishing and managing these services. Offering more than one model through a facility is likely to improve uptake.
Background: HIV clinics are struggling to absorb new patients in Malawi, and overburdened health-workers and long waiting times can be detrimental to adherence. We evaluated a strategy of six-monthly appointments (SMA) for stable ART patients in Chiradzulu District, Malawi, where Medecins sans Frontieres is supporting the Ministry of Health"s HIV programme.
Methods: Stable patients (aged ≥15, on first-line ART ≥12 months, CD4 count ≥300 and without opportunistic infections or ART intolerance, not pregnant or breastfeeding) were eligible for clinical assessments every 6 months instead of 1-2 months at 11 HIV clinics. Early SMA enrolees were defined as patients who started SMA within 6 months of eligibility, late SMA enrolees were those starting >6 months after eligibility.
Kaplan-Meier methods were used to calculate cumulative probabilities of death and loss to follow-up (LTFU) among those eligible for SMA, stratifying by SMA enrolment status and baseline characteristics. Cox regression, using SMA enrolment as a time-dependent variable, was used to estimate crude and adjusted hazard ratios for the association between SMA and death or LTFU.
Results: Between 2008 and 2015, 18,957 individuals were eligible for SMA (contributing 43,888 person-years of observation), of whom 15,308 (80.8%) ever enrolled. Median time from SMA eligibility to enrolment was 6 months (interquartile range 0-17 months). The cumulative probability of death or loss to follow-up five years after first SMA eligibility was 56.3% (95% CI: 52.4-60.2%) among those never SMA enrolled; 13.9% (95% CI: 12.5-15.6%) among early SMA enrolees and 8.1% (95% CI 7.2-9.0%) among late SMA enrolees.
After adjusting for age, gender, year of first SMA eligibility, and other baseline variables (CD4 count, months on ART and in cohort), a significantly higher rate of death or LTFU was observed among patients during non-SMA periods compared to those during SMA periods (adjusted rate ratio: 1.87, 95% CI 1.68-2.08, p< 0.001). Conclusions: SMA represents a promising strategy for managing stable ART patients and should be rolled out, particularly with “test and treat” on the horizon, which will further stretch HIV clinics. However, further implementation research is needed, and selection biases which may explain poor retention among those eligible but never SMA-enrolled should be investigated.
C Cawley, S Nicholas, E Szumilin, S Perry, I Amoros Quiles, C Masiku, A Wringe
Background: Long patient wait time is reported as an operational barrier to retention in HIV care in resource limited settings. Patients may perceive waiting several hours to see a clinician for only a few minutes as an unacceptable opportunity cost. The SEARCH HIV test-and-treat cluster randomized trial (NCT:01864603) in 32 rural Ugandan and Kenyan communities is implementing a “streamlined” HIV care delivery model in government supported clinics that aims to reduce wait times to address this problem.
Methods: We examined differences in patient wait time before and during clinical visits conducted under “streamlined” and standard government HIV clinic care. Components of streamlined HIV care aimed at reducing wait time included: (1) nurse-driven triage for patient evaluation; (2) 3-month ART refills (vs. 1 or 2 month) for stable patients; and (3) consolidation of services at encounter (ART, phlebotomy, medication dispensing). We conducted a time-and-motion study of patient clinical visits. We compared mean patient wait time before and during clinical visits among SEARCH study patients with CD4 >500 cells/uL (n=119), SEARCH patients with CD4 < 500 cells/uL (n=234) and other government clinic patients (n=745).
Results: Mean visit length was over one hour shorter among SEARCH patients with CD4 >500 cells/uL and SEARCH patients with CD4 < 500 cells/ uL compared to other government clinic patients, even though mean time with providers was similar between groups (see Table). This difference was due to wait times that were >30 minutes shorter both before and during visits. Time spent receiving health education, HIV care, laboratory services, medication dispensing and other services did not differ between patient groups.
Conclusions: Streamlined HIV care delivery led to shortened wait times both before and during HIV clinic visits. These efficiency improvements may contribute towards improved retention in HIV care.
SB Shade, W Chang, JG Kahn, D Mwai, F Mwangwa, D Kwarisiima, A Owaraganise, J Ayieko, DV Havlir, MR Kamya, ED Charlebois, ML Petersen, TD Clark, EA Bukusi, CR Cohen, V Jain
Background: The WHO advocates for differentiated HIV care and considers a broad range of community-based care models for patients stable on anti-retroviral therapy (ART). These care models aim to better respond to patient needs and to alleviate pressure on health systems caused by rapidly growing patient numbers. Most settings, however, utilized a single community-based care model only. We operationalize a combination of community ART care models in public health sector and assessed early outcomes.
Methods: Three community ART delivery care models were deployed in the rural Shiselweni region (Swaziland), from 02/2015 to 12/2015. First, Treatment Clubs (TC) are groups of 30 patients stable on ART who meet every 3 months at a secondary health facility for patient education and drug-refills. Second, Community ART Groups (CAG) comprise a maximum of 6 patients who alternate to attend the primary health clinic for consultation and pick up drugs for the other group members. Third, Comprehensive Outreach Care (COC) integrates drug refills into existing mobile clinic outreach activities for geographically isolated communities. We described baseline factors at enrolment,and 6 month retention in community care models and proportion of patients transferred back to routine clinical care.
Results: On average, 47 patients enrolled into community-ART care each month: 51.1% into TC
(242 patients in 8 groups), 34.0% in CAG (164 patients in 38 groups) and 14.9% in COC (65 patients in 2 remote communities). All patients had a VL< 1,000 copies/ml, the median CD4 was 512 (TC), 528 (CAG) and 657 (COC) cells/µl (p=0.27), the median age was 40, 40 and 45 years (p=0.11), and 74.8%, 66.5% and 64.6% were females (p=0.03). Retention in care after 6 months was highest in TC (97.5%) when compared to CAG (79.2%) and COC (78.4%) (p< 0.01). 53/471 patients (11.3%) returned back to and were retained in routine clinic care and one (0.21%) was recorded as death in COC.
Conclusions: Concurrent implementation of three community ART care models was feasible. Although a proportion of patients returned back to clinic care, overall ART retention was high and should encourage program managers to apply differentiated care models adapted to their specific setting.
L Pasipamire, B Kerschberger, I Zabsonre, S Ndlovu, G Sibanda, S Mamba, S Mazibuko, N Lukhele, SM Kabore, B Rusch